© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130390
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numerical expressions and statistics (counts, percentages, casualty figures, and similar number-based data).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    death and losses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     can
    -1.27
     nadat
    -1.20
    коменду
    -1.08
    It
    -1.04
     мощность
    -1.02
    شویی
    -1.02
    woordig
    -1.01
     urgently
    -1.01
     contactar
    -1.01
    atelli
    -0.99
    POSITIVE LOGITS
     losses
    1.48
     deaths
    1.43
     killed
    1.38
     loss
    1.34
    juries
    1.30
    alucía
    1.20
     death
    1.19
    owulf
    1.19
     nombreuses
    1.16
    atized
    1.13
    Activations Density 0.023%

    No Known Activations